The description of prior art will be primarily based on the list of references presented at the end of this section.
Concerns over the safety of earlier lithium secondary batteries led to the development of lithium ion secondary batteries, in which pure lithium metal sheet or film was replaced by carbonaceous materials as the anode. The carbonaceous material may comprise primarily graphite that can be intercalated with lithium and the resulting graphite intercalation compound may be expressed as LixC6, where x is typically less than 1. In order to minimize the loss in energy density due to this replacement, x in LixC6 must be maximized and the irreversible capacity loss Qir in the first charge of the battery must be minimized. Carbon anodes can have a long cycle life due to the presence of a protective surface-electrolyte interface layer (SEI), which results from the reaction between lithium and the electrolyte during the first several cycles of charge-discharge. The lithium in this reaction comes from some of the lithium ions originally intended for the charge transfer purpose. As the SEI is formed, the lithium ions become part of the inert SEI layer and become irreversible, i.e, they can no longer be the active element for charge transfer. Therefore, it is desirable to use a minimum amount of lithium for the formation of an effective SEI layer. In addition to SEI formation, Qir has been attributed to graphite exfoliation caused by electrolyte solvent co-intercalation and other side reactions [Refs. 1-4].
The maximum amount of lithium that can be reversibly intercalated into the interstices between graphene planes of a perfect graphite crystal is generally believed to occur in a graphite intercalation compound represented by LixC6 (x=1), corresponding to a theoretical specific capacity of 372 mAh/g. In other graphitized carbon materials than pure graphite crystals, there exists a certain amount of graphite crystallites dispersed in or bonded by an amorphous or disordered carbon matrix phase. The amorphous phase typically can store lithium to a specific capacity level higher than 372 mAh/g, typically up to 700 mAh/g, although a specific capacity higher than 1,000 mAh/g has been sporadically reported. Hence, the magnitude of x in a carbonaceous material LixC6 varies with the proportion of graphite crystallites and can be manipulated by using different processing conditions, as exemplified in [Refs. 1-4]. An amorphous carbon phase alone tends to exhibit a low electrical conductivity (high charge transfer resistance) and, hence, a high polarization or internal power loss. Conventional amorphous carbon-based anode materials also tend to give rise to a high irreversible capacity.
The so-called “amorphous carbons” commonly used as anode active materials are typically not purely amorphous, but contain some micro- or nano-crystallites with each crystallite being composed of a small number of graphene sheets (basal planes) that are stacked and bonded together by weak van der Waals forces. The number of graphene sheets varies between one and several hundreds, giving rise to a c-directional dimension (thickness Lc) of typically 0.34 nm to 100 nm. The length or width (La) of these crystallites is typically between tens of nanometers to microns. Among this class of carbon materials, soft and hard carbons made by low-temperature pyrolysis (550-1,000° C.) exhibit a reversible capacity of 400-800 mAh/g in the 0-2.5 V range [Refs. 1-3]. Dahn et al. have made the so-called house-of-cards carbonaceous material with enhanced capacities approaching 700 mAh/g [Refs. 1, 2]. Tarascon's research group obtained enhanced capacities of up to 700 mAh/g by milling graphite, coke, or carbon fibers [Ref. 3]. Dahn et al. explained the origin of the extra capacity with the assumption that in disordered carbon containing some dispersed graphene sheets (referred to as house-of-cards materials), lithium ions are adsorbed on two sides of a single graphene sheet [Refs. 1, 2]. It was also proposed that Li readily bonded to a proton-passivated carbon, resulting in a series of edge-oriented Li—C—H bonds. This provides an additional source of Li+ in some disordered carbons [Ref. 5]. Other researchers suggested the formation of Li metal mono-layers on the outer graphene sheets [Ref. 6] of graphite nano-crystallites. The amorphous carbons of Dahn et al. were prepared by pyrolyzing epoxy resins and may be more correctly referred to as polymeric carbons. Polymeric carbon-based anode materials were also studied by Zhang, et al. [Ref. 16] and Liu, et al. [Ref. 17].
Peled and co-workers improved the reversible capacity of a graphite electrode to 400 mAh/g by mild air oxidation [Ref. 4]. They showed that mild oxidation (burning) of graphite produces well-defined voids or nano-channels, having an opening of a few nanometers and up to tens of nanometers, on the surface of the graphite. They believed that these nano-channels were small enough to prevent co-intercalation of the solvent molecule but large enough to allow Li-ion penetration [Ref. 4]. These nano-channels were formed at the La-Lc interface, called “zigzag and armchair faces” between two adjacent crystallites, and in the vicinity of defects and impurities. Both natural and synthetic graphite materials typically have a wide variety of functional groups (e.g., carbonate, hydrogen, carboxyl, lactone, phenol, carbonyl, ether, pyrone, and chromene) at the edges of crystallites defined by La and Lc [Ref. 7]. These groups can react with lithium and/or electrolyte species to form a so-called in situ CB-SEI (chemically bonded solid electrolyte interface) [Ref. 4] on which, for example, carboxylic acid surface films are converted into Li-carboxylic salts.
Subsequently, several research groups have followed similar approaches to prepare mildly oxidized graphite anodes using a gaseous or liquid oxidant [Refs. 7-15]. For instance, Wu, et al. [Refs. 12-15] studied the electrochemical behaviors of natural graphite treated with (NH4)2S2O8, Ce(SO4)2, and H2O2. They similarly concluded that mild oxidation serves to (1) remove some active sites or defects in graphitic materials resulting in an improved surface structure; (2) form a dense layer of oxides acting as an efficient passivating film; and (3) produce nano-channels or micro-pores as storage sites and passages for lithium.
In summary, in addition to the above-cited three mechanisms, the following mechanisms for the extra capacity over the theoretical value of 372 mAh/g have been proposed [Ref. 4]: (i) lithium can occupy nearest neighbor sites; (ii) insertion of lithium species into nano-scaled cavities; (iii) in very disordered carbons containing large fractions of single graphene sheets (like the structure of a house of cards) lithium may be adsorbed on both sides of single layer sheets [Refs. 1, 2]; (iv) correlation of H/C ratio with excess capacity led to a proposal that lithium may be bound somehow in the vicinity of the hydrogen atoms (possible formation of multi-layers of lithium on the external graphene planes of each crystallite in disordered carbons) [Ref. 6]; and (vi) accommodation of lithium in the zigzag and armchair sites [Ref. 4].
It is of significance to emphasize that the approach of mild oxidation causes the formation of nano-channels at the zigzag and armchair faces between two adjacent crystallites and in the vicinity of defects and impurities [e.g., as stated in the Abstract of Ref. 4]. X-ray diffraction studies indicate that “the chemical oxidation of the graphite powder does not involve the formation of an intermediary graphite intercalation compound (GIC), since the interlayer spacing remains constant (3.354 Å)” [Second paragraph of the right column, Page 2969 of Ref. 7]. This implies that the bulk of the crystallites remains intact; only the amorphous phase, defects and impurities, and the interface between a crystallite and the amorphous phase have been modified. The reversible capacity of the resulting mild oxidation-treated graphite anode rarely exceeds 400 mAh/g [Refs. 4, 9] and, in most cases, still falls short of 372 mAh/g.